Prevalence and correlates of Mycoplasma genitalium infection among patients attending a sexually transmitted infection clinic in Guangdong, China: a cross-sectional study

Background Mycoplasma genitalium (MG) causes urogenital tract infections and is associated with reproductive morbidity. Although MG has been reported across many regions and population groups, it is not yet routinely tested for in China. Our study contributes to current research by reporting the prevalence and correlates of MG infection in patients attending a sexually transmitted infection (STI) clinic in Guangdong from Jan 2017-May 2018. Methods Urethral (from 489 men) and endo-cervical (from 189 women) samples, blood samples, and patient histories (via questionnaires) were collected. Doctors clinically diagnosed anogenital warts (GW) during the examination (n = 678). The presence of MG was evaluated using an in-house via polymerase chain reaction protocol. We also tested all participants for herpes simplex virus-2 (HSV-2), Neisseria gonorrhoeae (NG), Chlamydia trachomatis (CT), syphilis and HIV. Univariate and multivariate logistic regression were used to evaluate factors associated with MG. Results MG was detected in 7.2% (49/678) of the patients (men, 7.4%; women, 6.9%). The MG positivity rate was 14.2% among symptomatic patients, and 5.6% for asymptomatic patients, respectively. Only 36.7% (18/49) Mg positive patients were symptomatic. Among the MG-infected patients, 10.2% were co-infected with CT, 6.1% with NG, 8.2% with HSV-2, 4.1% with syphilis and 22.4% with GW. Presentation with clinical symptoms was significantly associated with MG infection [OR = 2.52 (2.03–3.13)]. In our analysis, MG was not associated with other STIs. Conclusions MG is a relatively common infection among individuals attending an STI clinic in Guangdong Province. Routine testing of symptomatic patients may be necessary, and more epidemiological studies are needed to provide evidence for future testing guidelines.


Conclusions: MG is a relatively common infection among individuals attending an STI clinic in Guangdong
Province. Routine testing of symptomatic patients may be necessary, and more epidemiological studies are needed to provide evidence for future testing guidelines.
Some existing epidemiological studies in China suggest that MG is widespread in China. In the available studies of MG infection in men living in China, the positive rate was 3.4% as detected in first-void urine (FVU) and 5.4% as detected in the rectal samples of men who have sex with men (MSM) in Shenzhen [14], 15.3% in the rectal samples of MSM in Shenyang [15]. Among men living with HIV in Jiangsu Province, the positive rate ranged from 20.1 to 29.2% as detected in FVU [16][17][18][19]. The positive rate of MG was 10.2% in symptomatic patients with NGU and 13.9% in patients with asymptomatic NGU in Hong Kong [20,21] and was 28.1% in men with STD-related complaints in Guangxi [22]. MG prevalence data for women are even more limited than those for men. However, one study reported a 13.2% prevalence in female sex workers living in Jiangsu [23].
In China, MG testing is not routinely performed in most sexually transmitted infection (STI) clinics, as there is no standard MG screening protocol to guide clinical practice, and commercial diagnostic kits were not available until 2019. National prevalence data for MG among routine STI patients are thus lacking. Therefore, we performed a cross-sectional study among patients attending a major urban STI clinic in Guangzhou, Guangdong Province, to assess the prevalence and correlates for MG infection.

Study design
From January 2017 until May 2018, we recruited patients attending our STI clinic in the Dermatology Hospital of Southern Medical University. Because our STI clinic is busy, we invited only patients who went to designated consulting rooms to participate in the study. Only first-time visitors were included in the study, these patients were symptomatic individuals seeking STI screening or asymptomatic individuals seeking STI screening due to high-risk sexual behaviours or having had contact with an infected individual. Patients aged less than 18 years or reporting any systemic antibiotic treatment in the previous 30 days were excluded. After the participants provided informed consent, their sociodemographic characteristics, sexual behaviour data and clinical findings were collected and documented as part of routine clinical procedure.

Sample size
The primary outcome of this study was the prevalence of MG. A previous study reported that the prevalence of MG was 11.2% [20]; thus, we calculated two-sided confidence intervals for one proportion to estimate a sample size of 651 for this study in order to produce a twosided 95% confidence interval with a width of 0.050.

Pathogen detection
Urethral swab specimens and FVU at least 2 h after the last previous urination were collected from all men, including those who reported no symptoms. The urethral swab specimens were used for urethral leucocyte count and MG detection. The urine specimens were tested for Neisseria gonorrhoeae (NG) and Chlamydia trachomatis (CT). Endo-cervical swab specimens were collected from women for MG/NG/CT nucleic acid testing and leucocyte count using microscopy. Blood sera were tested for HIV, Treponema pallidum (TP) and herpes simplex virus 2 (HSV-2) antibodies.
The amplification assay for MG detection was performed as follows: DNA was extracted using a commercial kit according to the manufacturer's instructions (Daan, Guangzhou, China). MG was detected via TaqMan MGB real-time polymerase chain reaction (PCR) with a sensitivity of five genome equivalents (geq)/reaction, as described by Jensen et al. [24]. Tests for NG and CT were performed using Roche's Cobas 4800 CT/NG test (Roche Diagnostics, Mannheim, Germany). Serum was tested for HIV using two antibody tests-a rapid HIV antibody test (Wantai, Beijing, China) and a second antibody test (Abon Biopharm, Hangzhou, China). If both were positive, another blood sample was collected and sent to the Guangzhou Center for Disease Control and Prevention for confirmation by Western blot analysis (MP Biomedical, Singapore). Serum was tested for syphilis using a toluidine red unheated serum test (TRUST, Rongsheng Bio-technology Limited Corporation, Shanghai, China) and Treponema pallidum particle agglutination (TPPA, Fujirebio Inc., Japan) test. Serum was tested for HSV-2 using the enzyme-linked immunosorbent assay (ELISA) method (Trinity Biotech, Guangzhou, China). Microscopic examination was performed on-site, and other samples were stored at − 20°C before testing.

Measures
Patients were considered symptomatic if they selfreported symptoms, such as dysuria, urethral pruritus, vaginal discharge, vaginal bleeding after sexual contact, and/or had abnormal findings during physical examination, such as urethral discharge (for males), abnormal vaginal discharge r yellowish mucopurulent t discharge at the cervical os and/or contact bleeding of the cervix (for females).
The urethral smear was regarded as presence of excess leucocytes when ≥5 polymorpho-nuclear leucocytes (PMNL) were seen per high power field (HPF) (× 1000) in ≥5 HPF, and the cervical smear was regarded as presence of excess leucocytes when > 30 PMNL were seen per HPF. Patients with positive TRUST and TPPA tests were considered to have an active syphilis infection unless there was a documented history of previously treated syphilis. Patients with positive HIV screening and confirmatory tests were considered to have HIV infection. Anogenital warts (GW) and anogenital herpes (GH) were diagnosed according to the diagnostic criteria for China's "National Notifiable Disease Reporting System" (Version 2008). These diagnoses were based mainly on exposure history and consistent clinical findings.

Ethics approval
The study was approved by the Ethics Committee of the Dermatology Hospital of Southern Medical University (approval no. GDDHLS-201502, 31/03/2015). The whole study including all its methods were carried out in accordance with relevant guidelines and regulations at the Dermatology Hospital of Southern Medical University, which are all in agreement with the Declaration of Helsinki. All data was anonymised prior to the analysis. Written informed consent was waived because the risk associated with participating in this study was deemed minimal and involved no procedures requiring consent outside of the context of participating in the study. Oral informed consent was obtained from all study participants prior to inclusion in the study. It is also in accordance with DHHS (45 CFR 46.117).

Statistical analysis
All data were entered into EpiData 3.0 via double entry for statistical analysis. Descriptive analyses were conducted on the patients' demographic characteristics and the prevalence of MG and other STIs. Categorical variables are presented as numerical and percentage values. Continuous variables are expressed as x AE SD values if the data were normally distributed and as median (P25 P75) values if the data were non-normally distributed. The chi-square test was used to compare categorical variables between MG-positive and MG-negative participants. Univariate analysis was used to evaluate factors associated with MG. A multivariate logistic regression model with stepwise variable selection was used to evaluate factors associated with MG based on the variables with a P-value of less than 0.1 in univariate analyses. For all analyses in this study, the results are reported as statistically significant when P ≤ 0.05. All data were analysed using SAS 9.4 (Statistics Analysis Systemint., Cary, NC, USA).

Characteristics of the study participants
During the 17-month study period, 727 individuals were recruited, 49 of whom were excluded due to missing data. A total of 678 patients-489 (72%) men and 189 (28%) women-were finally included in the present study. The mean age of the study population was 34.02 ± 13.39 years (33.70 ± 9.23 years for the men and 34.83 ± 20.35 years for the women. Of the 489 men, 41 (6%) identified as gay and 8 (1%) as bisexual. None of the women identified themselves as lesbian or bisexual. Half of the participants were married, and 66% had completed higher education. Most (77%) had been living in Guangzhou for more than 6 months.

Correlates of MG infections
Univariate logistic regression showed that symptomatic patients had a higher MG infection rate than asymptomatic patients (P = 0.001). Furthermore, patients who reported not using a condom during all sexual encounters were more likely to be infected with MG than patients who reported using a condom during each sexual encounter in the past 6 months (P = 0.035). Multivariable logistic regression showed that only clinical symptoms were associated with MG infection (Table 4). However, MG infection was not associated with other STIs, according to our analysis (Table 3).

Discussion
In our study, the prevalence of MG among patients attending our STI clinic was 7.2% (7.4% men, 6.9% women), and 36.7% of patients with MG infection were symptomatic. Among symptomatic patients, the prevalence of MG was 14.17% (18/127, 16 were men). Presentation with clinical symptoms was significantly associated with MG infection [OR = 2.52 (2.03-3.13)].
The strength of our study was that we tested a large number of STI attendees to provide data on the epidemiology of MG relative to other STIs, from a setting where MG testing is not routine. Thus, our data contributes to the sparse data on MG epidemiology, and could help in the development of national clinical guidelines for MG testing and screening protocols and the global efforts on MG monitoring. Limitations of our study included the inability to recruit larger numbers of female patients. Moreover, the patients were recruited from one hospital, and the results may thus not be generalizable to all patient populations in China.
The prevalence rates (7.2, 7.4%, respectively) among all patients and men attending our clinic were higher than *P6m refers to the 6 months prior to interview; ** refers to the definition of "symptomatic" made in the measures part in the text those among patients attending STI clinics with symptoms of urogenital tract infections or for a check-up in Greece (5.7, 6.4%, respectively) [25] and Norway (4.9, 4.5%, respectively) [26] but lower than the corresponding rates in Denmark (9.0, 8.9%, respectively) [26], Sweden (9.8, 9.1%, respectively) [26] and the US (16.7, 17.2%, respectively) [27]. The MG prevalence among men with symptoms were 14.2% (18/127) in our study, 12.3% in a study in London [28] and 16.7% in a study in Brussels [29] The MG prevalence in women in our study was 6.9% (95% CI: 3.2-10.5%), similar to the rates in women attending STI clinics with symptoms of urogenital tract infections or for a check-up in Greece (6.9%) [25] and Norway (6.0%) [26] but lower than the corresponding rates in Denmark (9.3%) [26], Sweden (11.1%) [26] and in two studies in the US (16.3, 17.5%) [27,30]. In our study, symptomatic patients were more likely to have MG infection than asymptomatic individuals (14.2% vs. 5.6%, P = 0.001). This pattern was also reported for men but not women among patients attending a young person's clinic in Sweden [31]. However, some studies arrived at the opposite conclusions, i.e., finding no statistical association between clinical symptoms and MG infection in men attending an STI clinic in Guangxi Province, China [22], and no association between MG infection and microscopically defined urethritis or cervicitis in patients in Greece [25]. These different conclusions may be related to the composition of the sample and the definition of symptoms.
In studies from Greece [25], Kenya [32] and Honduras [33], consistent condom use was not protective against MG. In our study, in the univariate analysis, patients who reported not using a condom during all sexual encounters in the past 6 months were more likely to be infected with MG than patients who reported using condom during each sexual encounter in the past 6 months (P = 0.035), while the multivariate analysis result failed to verify the linkage between protection of Mg infection and condom use.  Consistent with previous findings from the UK [28] and Greece [34], no association of MG infection with other STIs was found in our study via the chisquare test, indicating that MG might be an independent pathogen in the genital tract. The high level of MG and CT/NG co-infection suggests that screening and treating CT may not greatly impact MG infection, since azithromycin (1 g), the first-line drug for treating CT infection, appears to be a suboptimal choice for MG treatment.
In conclusion, MG was one the most common pathogen causing infection in individuals attending an STI clinic in Guangdong (7.2%). MG was an independent infection, and symptomatic patients were more likely than asymptomatic patients to be infected. Improving the accessibility of Mg testing in STI clinics in Guangdong Province is necessary, and more epidemiological studies are needed to provide evidence for future testing guidelines.